Talk:Perpetual check

old talk

I must have lost a point somehere, but isn't the knight on g5 ? How can it jump to g4 ? Is there aomething wrong somewhere ? (if what's wrong is in my head, I'd be glad to know what... ^^;; )

someone changed the diagram [1] Arvindn 19:39, 14 Oct 2004 (UTC)

historical perpetual-check rule

I have been Googling for a while and haven't found anything saying that "perpetual check used to be in the laws of chess". The best I could find was a mention by Staunton that it is a special case of the threefold repetition rule (which is the same status perpetual check has now - a tactic, not an independent drawing method). However, I haven't found a citable source saying there was never such a rule (just a lot of forum posts). I don't discount that whoever said there used to be a perpetual-check rule might have had some basis for saying that, so I am asking for a source citation.JoeJust 21:49, 22 March 2007 (UTC)

Official Chess Rulebook by Kenneth Harkness, 1970, page 46, says

...commonly known as the draw "by repetition of moves." It includes perpetual check

but it doesn't specifically list perpetual check. However, the way it is worded seems to imply that there had been a perpetual check rule. I'll look for older rulebooks. Bubba73 (talk), 22:16, 22 March 2007 (UTC)

I finally found a book old enough to have the perpetual check rule - a 1954 Reinfeld book. Citation added to article. Bubba73 (talk), 00:36, 3 April 2007 (UTC)

Additional information: on a chess newsgroup, someone said that the 1951 edition of "The Offical Blue Book and Encyclopedia of Chess" by Kenneth Harkness, page 51 states

the old rule covering draw by perpetual check has been abandoned. A player who can subject his opponent's King to an endless series of checks can force a third repetition of the position and claim the draw.

. However, I don't have a copy of that book to confirm it.

Also, the 1940 book How to play Chess Endings by Eugene Znosko-Borovsky, page 259 says "Thanks to perpetual check, Black succeeds in drawing the game." I think there is ample evidence that perpetual check was once a rule. Bubba73 (talk), 14:58, 11 April 2007 (UTC)

I wouldn't set much store on the Reinfeld quote. Explanations of the rules of chess outside chess rulebooks are often inaccurate in this kind of subtle detail. The Znosko-Borovsky quote has absolutely nothing to do with the question of whether perpetual check was ever a rule. For example, the same book probably says something like, Because he reaches a Philidor position, Black succeeds in drawing the game. But nobody would assume that this means that there is (or used to be) a rule that says that, if a Philidor position is reached, the game is automatically drawn. Dricherby 13:56, 4 June 2007 (UTC)

I'll rephrase it slightly. But it is a verfyable statement from a WP:RS. Bubba73 (talk), 15:29, 4 June 2007 (UTC)

I've ordered a copy of Kenneth Harkness's 1967 rulebook to see what it says. And note that above, someone told me that his 1951 rulebook says

the old rule covering draw by perpetual check has been abandoned. A player who can subject his opponent's King to an endless series of checks can force a third repetition of the position and claim the draw.

. Bubba73 (talk), 03:46, 7 June 2007 (UTC)

Now there is also the reference from Staunton. That sufficies for references. Bubba73 (talk), 14:46, 9 June 2007 (UTC)

It would be interesting to hear thge justification for removing the rule. I suspect that someone on the rules commitee raised the question of "How can one prove that something is 'endless' in a finite period of time?" It may be obvious to anyone and everyone, but proven? Not in the strict mathematical or logical sense. So, they let it go, as there was no practical effect: the draw would eventually be confirmed by one of the other rules. WHPratt (talk) 16:25, 9 February 2010 (UTC)

I think you must be right about that. In some cases perpetual check is obvious - in others it isn't so clear and there could potentially be a way out. The objective threefold and fifty-move rules will eventually be used, or a draw by agreement. I think that the FIDE rules committee wanted to make everything objective - no judging necessary. I think this applied to when a player's time runs out too. In FIDE rules, if one player's time runs out, the other must possibly be able to checkmate in order to win. In the case of Monica Socko it got down to K&N vs. K&N when one person's time expired. Checkmate is possible but the opponent has to help! They initially ruled a draw, but by the rules it was changed to a win. But no judgment is needed. Instead, the USCF has the "insufficient losing chances" rule, which makes more sense - except that it requires a judgement. Bubba73 ^{(You talkin' to me?)}, 18:04, 9 February 2010 (UTC)

Hamppe-Meitner

The article gives a slightly erroneous impression about this game, actually White's best outcome is a draw following 12.Kxc5 (moves following that other than those actually played lead to material disadvantage [13.Qg4?] or mate [14.Be8?? etc.]). As written in the article and in the caption for the position it seems to be implied that Black just manages to salvage a draw after White's 16th move. Mallocks 15:41, 19 September 2007 (UTC)

That was added by user:Krakatoa on Oct 28, 2006. Bubba73 (talk), 19:14, 19 September 2007 (UTC)

I've toned down the sheer amount of excited commenting but I'm still dubious as to whether the game should be included at all. Certainly it's beautifully played but the perpetual check at the end is just about the least exciting bit about it, so I question the particular relevence. Mallocks 11:03, 20 September 2007 (UTC)

Well, it does illustrate a perpetual check, so I'm in favor of keeping it. Bubba73 (talk), 15:47, 20 September 2007 (UTC)

I've tried to clear it up. Mallocks 17:19, 20 September 2007 (UTC)

Examples

The example pic with the queen is wrong. The black king can block with a pawn move to G6, stopping perpetual check. The pic needs to be altered/removed/replaced. —Preceding unsigned comment added by 24.4.178.155 (talk) 03:41, 8 December 2007 (UTC)

Agreed. 160.39.226.24 (talk) 03:46, 24 April 2008 (UTC)

Example cut: not necessarily perpetual check

I cut this example from the article:

	a	b	c	d	e	f	g	h
8									8
7									7
6									6
5									5
4									4
3									3
2									2
1									1
	a	b	c	d	e	f	g	h

Black to move can draw by perpetual check

In the second position, Black draws by giving perpetual check:

1... Qa1+!

2. Kd3 Qd1+

3. Ke3 Qg1+!

4. Ke4 Qg4+

5. Kd5 Qd7+

6. Kc5 Qa7+

7. Kc4 Qa4+

8. Kc3 Qa1+, etc.

By checking on the marked squares (including a4), White has no chance to stop the checks (Speelman 1981:34–35).

This is a draw but not necessarily a draw by perpetual check; white can avoid the checks and still draw the game by a sequence like this:

1... Qa1+

2. Kc2 Qxd4

3. g8=Q

You can confirm this by entering the original position (8/6P1/8/8/q2Q4/2K5/8/7k b - - 0 0) to the Nalimov Tablebase server. Gdr 19:47, 26 September 2008 (UTC)

Yea, well... White either has to give in to a perpetual check or give away his queen, and obviously he doesn't want to do that. Remember it is Black that is trying to draw, not White. Bubba73 (talk), 05:07, 27 September 2008 (UTC)

Is this a normal meaning of "perpetual check"? That is, it doesn't mean "inescapable sequence of checks" but "sequence of checks that can only be escaped with disadvantage?" If this is the case, then the example could be restored to the article with an explanation and a reference to a chess writer who uses the phrase with this meaning. Gdr 11:39, 27 September 2008 (UTC)

I'll look up some precise definitions later and see, But here White can escape the perpetual check by giving up his queen. But since White wants to play for a win, he obviously doesn't want to do that. Bubba73 (talk), 16:20, 27 September 2008 (UTC)

The Oxford Companion to Chess simply says "an unstoppable series of checks usually instigated to avoid loss of the game", which doesn't clear it up. Golombek's Encyclopedia isn't any more helpful. But I think it is implied that if he has to give up his advantage (or even get checkmated) to get out of the perpetual check, it is still basically perpetual check. Bubba73 (talk), 02:20, 29 September 2008 (UTC)

The only way White can escape the checks is by hanging his queen, resulting in a dead-drawn Q v. Q ending after Black takes White's queen and White promotes her pawn to a new queen. I think under those circumstances it's reasonable to describe the position as one where Black draws because of perpetual check. Krakatoa (talk) 03:58, 29 September 2008 (UTC)

Yes, in this case I think it is safe to say that the perpetual check forces a draw, one way or the other. I don't think Black can draw without it. Bubba73 (talk), 04:55, 29 September 2008 (UTC)

No-one doubts, I think, that this position has perpetual-check-like features. I have three questions: (1) Can we verify that chess writers use the term "perpetual check" to include positions like this? In particular, what does Speelman say in the cited work? (2) How common is it for chess writers to use the term broadly? Is it a majority or minority position? Can we find examples other than Speelman? (3) Even if we can find several chess writers who use the term broadly, is it a good idea for pedagogical reasons to place this example second in the article? Might it not be better towards the end, with an introduction along the lines of "some chess writers, e.g. X and Y, define the term more broadly, including positions where a sequence of checks can be escaped, but the escape leads to disadvantage for the escaping player." Gdr 12:51, 29 September 2008 (UTC)

For #1, Speelman has a section "Perpetucal check and perpetual attack". He says "In a sense perpetual check is only one example of perpetual attack - any manoeuvre whereby the weaker side forces the stronger side to repeat a position. First some examples of perpetual check." Then he gives that position, with the caption "example of perpetual check". Then he continues "Perpetual check, or the threat of it, is one of the defender's main resources in queen and pawn endgames. In diagram 60 Black can give perpetual check immediately..." and he gives moves. "By checking on the squares which mark out the perimeter of a square Black gives White no chance to vary." He says nothing about giving up the white queen to end the perpetual check. But even if White does this, Black can either have a perpetual check or force the exchange of queens. Chess writers make some assumptions about what the reader will understand. Bubba73 (talk), 15:10, 29 September 2008 (UTC)

What if we change it to "White cannot stop this series of checks except by giving up his queen." Bubba73 (talk), 15:16, 29 September 2008 (UTC)

Yes, that sounds like a good way to phrase it. Gdr 15:28, 29 September 2008 (UTC)

first example

Moved from the article page: "This is for anyone who looks at this page, but if I'm not mistaken, when the Queen moves to F5, the pawn at G7 can just move to G6, blocking the Queen and actually preventing perpetual check. Would anyone like to provide a better example of perpetual check? "

Black does move the pawn to g6 on his fourth move to stop the checks, but then the white queen captures the pawn on f7 with check, perpetual check. Perhaps the example should show the pawn already on g6. Bubba73 (talk), 20:43, 21 January 2009 (UTC)

Unzicker versus Averbakh

h in step 3 should be p for pawn--L

Nope. "P" for pawn is not used in algebraic notation; specifying the starting file as done here is the correct way to notate a pawn capture. Double sharp (talk) 11:30, 5 April 2019 (UTC)

Perpetual pursuit

	a	b	c	d	e	f	g	h
8									8
7									7
6									6
5									5
4									4
3									3
2									2
1									1
	a	b	c	d	e	f	g	h

Perpetual pursuit

We should probably give an example of perpetual attack / perpetual pursuit that is not check, come to think of it. A nice study is this one by S. Birnov (1928) given in Seirawan's Winning Chess Tactics). 1.a6 Bxc4 2.e4+! (restricting the bishop to half the diagonal) 2...Kxe4 3.a7 Bd5 4.c4 Ba8 5.Kb8 Bc6 6.Kc7 Ba8 and the perpetual pursuit is achieved. (An immediate 1.e4+? fails to 1...Ke5! 2.a6 Bh7 3.a7 Bxe4 and the pawn is stopped.) Double sharp (talk) 08:56, 2 October 2021 (UTC)

(Bilek vs Schussler 1978 is a real-game example.) Double sharp (talk) 12:25, 2 October 2021 (UTC)

I think that is OK. It probably needs a little explanation of what will happen from there in the 1978 game. Bubba73 ^{You talkin' to me?} 16:26, 2 October 2021 (UTC)

I added it. Strangely Seirawan and Silman don't attribute the Bilek vs Schussler game, so I should probably find a source for that. Double sharp (talk) 08:53, 5 October 2021 (UTC)

Chekhover and Korolkov, Chigorin Memorial, 1949–1950

	a	b	c	d	e	f	g	h
8									8
7									7
6									6
5									5
4									4
3									3
2									2
1									1
	a	b	c	d	e	f	g	h

Reciprocal pursuit

P.S. Also possible is reciprocal pursuit when both sides have nothing better than to chase each other. See the study to the right. I'd like to add it (probably without the lead-in), if I can find an actual sourced annotation for what's going on. Mainline from Roycroft's The Chess Endgame Study: A Comprehensive Introduction (it's #375 there): 1.f7 Bd6+ 2.Kb2 Nh6 3.f8=Q+ Bxf8 4.Nxf8 Rf4 5.Ng4 Nxg4 and the pursuit starts: 6.Ne6 Re4 (6...Rf6 7.Nc7+ Kb7 8.Nd5 and 9.hxg4) 7.Ng5 (7.Nc5? Rxc4 8.Nd7 Rd4 9.Nb6+ Kb7 10.Kc3 Rd8 11.Nc4 Nh6; 7.Nc7+? Kb8 8.Na6+ Ka7 9.Nc5 Rxc4 and wins) 7...Re5 8.Nf3 Rf5 9.Nd4 Rf4 10.Ne6 Re4 11.Ng5 and drawn. Double sharp (talk) 09:04, 5 October 2021 (UTC)

Larry Kaufman's Kaufman's New Repertoire for Black and White uses "perpetual 'check' to the queen" to refer to such a perpetual pursuit in his opening analysis of the Marshall: 1.e4 e5 2.Nf3 Nc6 3.Bb5 a6 4.Ba4 Nf6 5.0-0 Be7 6.Re1 b5 7.Bb3 0-0 8.c3 d5 9.exd5 Nxd5 10.Nxe5 Nxe5 11.Rxe5 c6 12.d4 Bd6 13.Re1 Qh4 14.g3 Qh3 15.Qf3 Bg4 16.Qg2 Qh5 17.Be3 Rae8 18.Nd2 Re7 19.a4 Rfe8 20.axb5 axb5 21.Bxd5 cxd5 22.Qf1 f5 23.f3 Bh3 24.Qxb5 f4 25.Bf2 fxg3 26.hxg3 Bd7 27.Qf1 Bh3 28.Qd3 Bf5 and there is nothing better than to repeat. Double sharp (talk) 23:22, 29 April 2023 (UTC)